March 7, 2014

This Week in HPC News

Nicole Hemsoth

March kicked off rather quietly compared to the rest of the year thus far, but with some interesting news items from DDN, SGI, and others—all of which we’ll address in a moment.

The break in an otherwise active 2014 news cycle gave us some time to reflect on some end user and other community issues via the daily podcast series, including training the next generation of HPC professionals. We also checked in about the emerging Big Data Top 100 benchmark, got a sneak peek ahead at SC14 with Conference Chair, Trish Damkroger, and talked about simulating galaxies and scaling gravity on systems at TACC and NASA Ames.

Before we move into the week’s top news, we wanted to take a moment and say congrats to the IEEE Technical Achievement Award Winners, who were selected across a number of areas ranging from computer security and parallel processing to pattern recognition and information. The following have been named as recipients:

Kevin Bowyer, the Schubmehl-Prein Professor and Chair of the Department of Computer Science and Engineering at the University of Notre Dame, for “pioneering contributions to the science and engineering of biometrics”;

W. Bruce Croft, a Distinguished Professor in the School of Computer Science at the University of Massachusetts Amherst, for “outstanding contributions to information retrieval and the development of search engines”;

Srinivas Devadas, the Webster Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology, for “pioneering work in secure hardware, including the invention of Physical Unclonable Functions and single-chip secure processor architectures”;

Pedro Felzenszwalb, associate professor of engineering and computer science at Brown University for “the deformable parts model method of detecting objects in images and video,” and

Albert Zomaya, the Chair Professor of High Performance Computing and Networking and Australian Research Council Professorial Fellow in the School of Information Technologies, The University of Sydney, “for outstanding contributions to the solution of scheduling problems in parallel and distributed computing systems.”

More Top News Items

SGI_logo_platinum_smlSGI Delivers Shared Memory System to Japan. The agency for Marine-Earth Science and Technology (JAMSTEC) has selected SGI UV 2000 powered by Intel Xeon processor E5-4600 v2 series for pre and post-processing offline grid simulations for its Earth Simulator supercomputer. Working with NEC, the SGI solution will be installed at the JAMSTEC Earth Simulator supercomputer center and is targeted for production on March 24th, 2014.

The JAMSTEC Earth Simulator supercomputer is integrated with the Intel Xeon processor E5-4600 v2, which Intel officially launched recently. The system will provide the center with, ” super fine-grained simulation results and dramatically increases compute efficiency. The solution will accelerate a number of research projects involving earth sciences, especially those related to global warming projections. In addition, the solution will be widely utilized by automotive manufacturers and both the pharmaceutical and chemical industries through a ‘Strategic Use Acceleration Program of Earth Simulator,’ a program designed to increase industrial usage through strategic partnerships.”

skaMellanox Speeds SKA via a collaboration with the University of Cambridge for the Square Kilometer Array (SKA) project. They selected the company’s Virtual Protocol Interconnect (VPI) solution, consisting of ConnectX-3 adapter cards, SwitchX-2 based SX1036 36-port switches and cables, to provide it with leading interconnect performance and protocol flexibility for SKA test-bed clusters. The University of Cambridge and Mellanox will use the compute clusters for various development projects for the SKA project, an international effort to build the world’s largest radio telescope.

“The computing requirements of the SKA exceed those of the fastest supercomputers, as the data processing and amounts of data generated by combining the signals from our antennas compete with that generated by the entire Internet,” said Dr. Paul Calleja, director HPCS Cambridge University. “Utilizing Mellanox’s protocol-flexible VPI solutions in the University of Cambridge supercomputer, we are able to run our applications over either Ethernet or InfiniBand on a single-wire to enable best application performance without the need for multiple networks or topologies.”

Hitachi Opens Big Data Lab in Demark, which is set to open by the end of September 2014. The goal of the laboratory is to use big data to create new information services that will help to realize a sustainable, “more comfortable” society.

Hitachi says it is expanding the Social Innovation Business globally. In Denmark, Hitachi has been expanding its business particularly in the Information & Telecommunication Systems sector of the Social Innovation Business, mainly in storage solutions and consulting services. Hitachi is also accelerating initiatives in big data related fields. Hitachi opened the Big Data Research Laboratory in the U.S. and in the U.K., respectively in April 2013 and in June 2013.

Clemson University has just been on the receiving end of a $5.3 million NSF grant to enable a national network of Advanced Cyberinfrastructure Research and Education Facilitators (ACI- REFs) to broaden the impact of advanced computing resources at campuses across the country.

The project, called the Advanced Cyberinfrastructure – Research and Educational Facilitation: Campus-Based Computational Research Support, is a consortium that brings together education and research institutions that are committed to the vision of advancing scientific discovery through a national network of Advanced Cyberinfrastructure (ACI) Research and Education Facilitators (ACI-REFs). Working together in a coordinated effort, the consortium is dedicated to the adoption of models and strategies to leverage the expertise and experiences of its members to maximize the impact of investment in research computing

Until next week—thanks for reading and stay tuned for more HPC news and analysis.

Share This